Background. Cannabinoids, initially thought to act through non-specific interactions related to cell membrane fluidity, were shown to activate cannabinoid receptors, cloned in the early 90s. Two types of cannabinoid receptors exist, CB1, most expressed in the brain, and CB2 mostly expressed in the periphery, but cannabinoids have been shown to activate other receptors, for example, TRPV1, PPAR, and GPR55. Different endogenous compounds bind to cannabinoid receptors, among them, anandamide (AEA), and 2-arachidonyl-glycerol (2AG), that are not stored in vesicles, like other transmitters, but released on demand by cleavage of membrane precursor components. The recent availability of natural and synthetic ligands, and drugs that block the metabolism or the uptake of endocannabinoids, has facilitated the study of the role of cannabinoids in physiological functions, for example in synaptic plasticity. Dysfunctions of the endocannabinoid system could also result in neurological disorders, including drug addiction. Endocannabinoid Neurochemistry. In collaboration with Drs. Goldberg and Solinas we have shown that the endocannabinoids AEA (also its metabolically stable analogue methAEA) and 2AG share with drugs abused by humans the ability to acutely stimulate DA levels in the rat NAC shell, suggesting endocannabinoids may possess reinforcing effects. Indeed, AEA promotes and maintains intravenous self-administration behavior in monkeys. These effects are blunted by pretreatment with a selective CB1 antagonist, indicating a direct CB1 receptor involvement in elevation of DA and reinforcing effects. Compounds known to increase brain endocannabinoid levels may also elicit DA release. URB597 inhibits the enzyme Fatty Acid Amide Hydrolase (FAAH) that metabolizes AEA. This drug alone (at doses that fully block FAAH) did not produce DA stimulation, suggesting that, in animals at rest, levels of AEA are not high enough to activate CB1 receptors, even after FAAH blockade. Alternatively, FAAH might not be sufficiently expressed in areas related to DA transmission. However, URB597 enhanced the effects of exogenous AEA on DA levels, supporting the hypothesis of low levels of AEA in animals at rest, and suggesting that URB597 can block AEA metabolism resulting in increased levels of AEA in areas related to DA transmission. URB597 alone does not elevate DA levels, and its degree of abuse liability appears low, in agreement with negative findings of self-administration in monkeys, and place preference in rats. AM404, another endocannabinoid enhancer, produces cannabinoid-like behaviors in rodents, but in our experiments it did not alter DA levels, and surprisingly, did not enhance the effect of AEA on NAC shell DA. In agreement, in rats trained to discriminate THC from saline (a behavior selectively mediated by CB1 receptors), AEA alone did not produce full THC-like effects (likely due to its rapid metabolism), while after pretreatments with URB597, but not with AM404, AEA produced full THC-like effects. DA D2 receptor activation stimulates AEA levels, thus, drugs able to increase DA levels in the brain, like drugs of abuse, might activate DA D2 receptors, and, in turn, increase AEA levels. Because we do not possess a technique to directly measure brain endocannabinoid levels, we have tested some drugs abused by humans in producing CB1-receptor mediated generalization in THC-discrimination tests. Cocaine, and amphetamine injected alone did not produce effects significantly different from vehicle, but potentiated the THC-like effects of ineffective doses of THC. Nicotine and the D2-D3 DA receptor agonist quinpirole administered alone did not generalize to the THC cue, but both drugs did so in animals pretreated with the inhibitor of FAAH that metabolizes AEA. Nicotine and quinpirole also potentiated the effects of THC. We have suggested that AEA is released by these drugs in specific brain areas by a D2 receptor mediated mechanism. So, administered alone these drugs do not stimulate AEA levels high enough to provide CB1-mediated THC-like effects, but potentiate ineffective small doses of THC. However, when the same drugs are administered in combination with URB597, AEA levels are magnified by blockade of its metabolism, and its concentration could thus activate CB1 receptors producing THC discriminative effects. Endocannabinoids and Psychostimulant Behavioral Sensitization. Psychostimulant-induced behavioral sensitization might play a role in the path to abuse and addiction. Sensitization can be viewed as a type of synaptic plasticity, which is also related to alterations in the cannabinoid system. We hypothesized that development of psychostimulant sensitization involves changes in endocannabinoid levels and their activity at CB1 receptors. We have started this project studying cocaine sensitization in mice, as cocaine has also been the focus of other projects in our branch. We are testing the hypothesis that the development of cocaine sensitization after a single cocaine exposure requires release of endocannabinoids, and can be reversed by CB1 antagonists. We are also testing if doses of cocaine that do not induce behavioral sensitization, might become effective when animals are pretreated with endocannabinoid level enhancers. DA transmission, believed to mediate behavioral and reinforcing effects of cocaine, will also be measured before and after cocaine behavioral sensitization in order to relate changes in brain neurochemistry to occurrence of behavioral sensitization. Preliminary results from this study have confirmed that a single exposure to a moderate/high dose of cocaine, 20 mg/kg i.p., induces behavioral (locomotor) sensitization in mice re-exposed to cocaine, 10-20 mg/kg i.p., 24 hours thereafter. Rimonabant, a selective CB1 antagonist, injected before the sensitizing dose of cocaine, blunted the behavioral sensitization produced by cocaine. Studies on NAC shell and core levels of DA in mice, to evaluate neurochemical changes related to behavioral sensitization are in progress. Role of Cannabinoid Receptors in Reinforcment Value Studied by Behavioral Economic Tests in Mice. Given the involvement of cannabinoid CB1 receptors in different aspects of drug abuse and addiction, we are investigating the role of CB1 receptors on the effectiveness of food reinforcement. This study, in collaboration with Drs. Lupica and Hoffman, is being performed in genetically modified CB1-KO mice, HET-mice and their wild-type littermates, using behavioral economic procedures. Under these procedures, mice kept at about 80 % of their typical weight are required to work (emit operant responses) to get a reward. As the price (number of operant responses mice have to make) is increased, consumption decreases. This demand function is derived from behavioral economic theory, and has been validated as a behavioral index of reinforcing efficacy. Differences between genotypes in their responding for rewards, and in demand curves, can be used as an index of differences in the value of a specific reinforcer. Studies are underway to evaluate the role of CB1 receptors in the value of specific reinforcers using these genetically engineered mice.